J. Phys. Chem. B 2007, 111, 10665-10670
10665
Manipulating Formation and Drug-Release Behavior of New Sol-Gel Silica Matrix by Hydroxypropyl Guar Gum Guan-Hai Wang and Li-Ming Zhang* Laboratory for Polymer Composite and Functional Materials, Institute of Optoelectronic and Functional Composite Materials, School of Chemistry and Chemical Engineering, Sun Yat-Sen (Zhongshan) UniVersity, Guangzhou 510275, China ReceiVed: January 16, 2007; In Final Form: June 19, 2007
To develop biocompatible sol-gel silica matrix for the encapsulation of biomolecules or drugs, a novel watersoluble silica precursor, tetrakis(2-hydroxyethyl)orthosilicates (THEOS), was used in combination with a water-soluble polysaccharide derivative, hydroxypropyl guar gum (HPGG). We found that the introduction of HPGG could trigger and accelerate the sol-gel transition of THEOS in water and induce rapid formation of homogeneous gel matrix without the addition of any organic solvents or catalysts. Moreover, added HPGG macromolecules had a great influence on the network structure and particle dimension in the silica gel matrix, as confirmed by scanning electron microscope (SEM) observation. From the time sweep rheological measurements, it was found that a higher HPGG amount could lead to shorter gelation time for the sol-gel transition. From the strain and frequency sweep rheological experiments, it was found that the resultant silica matrix containing a higher amount of HPGG exhibited a narrower linear viscoelastic region, a higher dynamic muduli, and greater complex viscosity. In particular, the gel strength of the silica matrix could be modulated by the amount of HPGG. By investigating the controlled release of vitamin B12 from the sol-gel silica matrixes, a strong dependence of the release profile on the amount of introduced HPGG was observed. In this case, a higher HPGG amount resulted in lower release rate.
Introduction Sol-gel silica matrixes are amorphous, porous, and biocompatible materials and have found many applications in enzyme immobilization, nanocomposite catalysts, biosensors, nonlinear optics,andthecontrolledproductionofsemiconductornanoparticles.1-7 In recent years, they have been also investigated for the potential uses as the carrier materials for controlled drug delivery.8-20 The aqueous character of the sol-gel process and the fact that the preparation of the silica materials can be conducted at roomtemperature ensure that sufficiently mild conditions prevail, as required for drug molecules to retain their native structure and functions. In this case, drugs can be trapped within the silica networks simply by mixing them with the silica precursor in the sol form. After the gelation, the drugs in silica sols become uniformly distributed within the porous silica gel networks. Moreover, these materials cause no adverse tissue reactions and can be degraded in the body to silicic acid, which may be eliminated through the kidney.21 In addition, they can provide enhanced mechanical properties in an economically and biologically safe manner when compared with pure polymeric matrixes. Although sol-gel silica materials have many advantages for the controlled release of drugs, there still exist some disadvantages: (1) Most of the used silica precursors do not dissolve enough in water. In order to get a uniform sol, an organic solvent is usually added, which is unfavorable for the encapsulation of water-soluble drugs and biocompatibility. (2) In most cases, the gelation time for the sol-gel transition is usually very long, especially when an organosilane is added to improve the textural properties of the gels and to adjust the interactions between the * To whom correspondence should be addressed. E-mail: ceszhlm@ mail.sysu.edu.cn.
drug and the gel matrix. (3) When a dopant such as a surfactant is used to prepare the gels for further modulation, some of the dopant may release from the gels together with the drug, which would affect the release profile of the drug embedded in solgel silica materials and its activity. In this study, we used a completely water-soluble silica precursor, tetrakis(2-hydroxyethyl)orthosilicates (THEOS), to fabricate new sol-gel silica materials for the controlled drug release. For this purpose, a water-soluble polysaccharide derivative with good biodegradability and biocompatibility, namely hydroxypropyl guar gum (HPGG),22,23 was used, as shown in Figure 1. By the control of HPGG amount, we found that the formation of such a sol-gel silica matrix and its drug-release behavior could be modulated effectively without the use of any organic solvents or surfactants. Experimental Section Materials. The water-soluble silica precursor, THEOS, was prepared from tetraethoxysilane (Shanghai Chemical Co., China) and glycol (Guangzhou Chemical Co., China) according to the method reported by Mehrotra and Narain.24 The HPGG used is a commercial product, which was kindly provided by China Agency Office of Economy Polymers & Chemicals Company. Its moisture content and the pH value of a 1% aqueous solution (25 °C) were determined to be 8.0% and 6.0, respectively. By using an Ubbelohde capillary viscometer, the intrinsic viscosity of HPPG was determined to be 931.5 mL/g at 30 ( 0.02 °C according to the Huggins equation.25 The model drug, vitamin B12, was purchased from Shanghai Boao Biomedical Company in China. The other chemicals were of analytical grade and used without further purification.
10.1021/jp070370a CCC: $37.00 © 2007 American Chemical Society Published on Web 08/21/2007
10666 J. Phys. Chem. B, Vol. 111, No. 36, 2007
Wang and Zhang
Figure 1. Structure of the repeating unit of hydroxypropyl guar gum (HPGG).
Preparation and Characterization of the New Silica Matrix. The new silica matrix was prepared by a one-step solgel process triggered by HPGG. A required amount of HPGG was initially dissolved in water and left for 24 h. After the addition of an appropriate weighed amount of THEOS, the mixture system was thoroughly stirred and set aside at ambient temperature. The gelation process was monitored by the rheological method,26 and the viscoelastic properties of the resulting gel were measured at room temperature by means of an advanced rheometric extended system (ARES, TA Co.) in oscillatory mode with a cone-and-plate geometry (cone diameter, 25 mm; angle, 1°). To ensure the rheological measurements within a linear viscoelastic region, a dynamic strain sweep was conducted prior to the frequency sweep, and the corresponding strain was determined to be 1.0%. Scanning electron microscope micrographs were taken by a JSM-6330F field emission scanning electron microscope. Before the SEM observation, the gel sample was frozen in a Virtis Freeze Drier (Gardiner, NY) under vacuum at -40 °C for at least 3 days and then fixed on an aluminum stub and coated with gold. For a comparison study, the sol-gel process of THEOS was also investigated in the absence of HPGG. Drug Entrapment and Release in the Sol-gel Silica Matrix. For the entrapment of the model drug, a required amount of vitamin B12 was first mixed with aqueous HPGG solution, and the sol-gel process was then carried out in the presence of THEOS until the formation of the silica gel. To investigate the drug release behavior, the silica gel loaded with vitamin B12 was placed in 60 mL of phosphate-buffered saline (PBS, pH 7.4). At different intervals, the aliquot of the solution was taken for the determination of the drug. A UV-2910 spectrophotometer (Shanghai Instrument Company, China) was used to determine the concentrations of the drug at 276 nm. All the release experiments were carried out in duplicate. Results and Discussion Formation of the Sol-gel Silica Matrix. When the silica precursor THEOS was dissolved in water, a low-viscous liquid was formed. It is known24 that THEOS in water could hydrolyze into glycol and silicic acid and then condense into the silica gels. However, we did not observe an obvious viscosity change or any gelation phenomenon for an aqueous solution of 10 wt % THEOS during the initial 24 h. In contrast, the rapid gelation could be observed after mixing THEOS with HPGG in aqueous solution. Moreover, phase separation or precipitation was not observed for such system. These phenomena demonstrate that HPPG could trigger and accelerate the sol-gel transition of THEOS in water. It is worth mentioning that such gelation could be induced without the addition of any organic solvents, which was different from the situation that occurred in the case of
Figure 2. The elastic modulus (G′) and viscous modulus (G′′) as a function of time for pure THEOS system and THEOS/HPGG hybrid system. Time corresponding to the crossover point between G′ and G′′ curves is taken as the gelation time. The experiments were carried out at 1 Hz, 1.0% strain, and 25 °C.
common silica precursors such as tetramethoxysilane (TMOSE) or tetrathoxysiliane (TEOS).27 To determine the rate and extent of the gelation, a time sweep measurement was carried out for each system, in which the storage modulus (G′) and loss modulus (G′′) were monitored as a function of time. Figure 2 shows the time dependence of G′ and G′′ for aqueous 10 wt % THEOS system and its mixed system with 1.5 wt % HPGG. For aqueous 10 wt % THEOS system, it was difficult to measure the G′ and G′′ values because of its low viscosity, and little change could be detected during the investigation (Figure 2a). When 1.5 wt % HPGG was introduced into aqueous 10 wt % THEOS system, however, the G′ and G′′ of the aqueous THEOS/HPGG system could be measured exactly and was observed to have an obvious increase with the increase of test time (Figure 2b). In particular, a crossover point between G′ and G′′ was observed for the aqueous THEOS/HPGG system, which implied that there was a sol-gel transition.26 Beyond the crossing, the G′ value becomes larger than the G′′ value, indicating that the system becomes more elastic. The corresponding time of the crossover from a viscous behavior to an elastic response could be regarded as the gelation time.26 From Figure 2b, three main regions could be distinguished in the time sweep curves. First, a short induction period was observed, where the G′ and G′′ values did not change appreciably with time. In the second region, the G′ and G′′ values increased
Sol-Gel Silica Matrix
Figure 3. Effect of HPGG amount on the gelation time for aqueous THEOS/HPGG system. The experiments were carried out at 1 Hz, 1.0% strain, and 25 °C.
sharply with time, inferring that the network structure began to form. This was followed by the third region, where the slopes of the G′ and G′′ curve versus time continuously decreased, inferring that the network formation process was reaching completion. Other THEOS/HPGG systems with different HPGG amounts exhibited similar gelation kinetic curves. Figure 3 gives the gelation time for all THEOS/HPGG systems investigated. As seen, an increase of the HPGG amount results in a decrease of the gelation time, which confirms the accelerating affect of HPPG on the sol-gel process of THEOS. In general, the silicabased sol-gel process occurs in three main stages:28,29 (1) Hydrolysis of the silica precursor:
Si(-OR)4 + nH2O f (HO-)nSi(-OR)4-n + nHO-R where R is a hydrocarbon radical and n e 4. (2) Condensation of the formed monomers to produce the oligomers arranged as sol particles:
2 (HO-)nSi(-OR)4-n f (OH)n-1(RO-)4-nSi-O-Si (-OR)4-n(OH)n-1 + H2O or
Si(-OR)4 + (HO-)nSi (-OR)4-n f (RO-)3Si-O-Si (-OR)4-n(OH)n-1 + HO-R (3) Cross-linking of the sol particles leading to a sol-gel transition. When HPGG was introduced into the precursor solution, the sol-gel process might be modified by the following condensation reaction:
(HO-)nSi (-OR)4-n + HO-HPGG f (OH)n-1(RO-)4-nSi-O-HPGG + H2O In this case, the added HPGG will have a catalytic effect on the sol-gel process of THEOS, which results in the shortening of the time interval for the gel formation. To understand the internal structure of the formed silica gel in the presence of HPGG and perform dynamic rheological characterization, the linear viscoelastic (LVE) regions were determined by strain sweep tests for all THEOS/HPGG systems studied, in which dynamic rheological parameters are indepen-
J. Phys. Chem. B, Vol. 111, No. 36, 2007 10667
Figure 4. The strain dependence of storage modulus at a frequency of 1 Hz for aqueous THEOS/HPGG systems (25 °C).
dent of applied strains. Figure 4 shows the strain dependence of the storage modulus at a frequency of 1 Hz for each system. The limit of the LVE region was found to be about 24% for the silica gel containing 0.2 wt % HPGG, 13% for the silica gel containing 0.5 wt % HPGG, 10% for the silica gel containing 1.0 wt % HPGG, and 8% for the silica gel containing 1.5 wt % HPGG, respectively. These results indicated that the extent of the LVE response was dependent on the HPGG amount. Moreover, a higher HPGG amount could lead to a reduction in the critical strain (γc) separating the linear and the nonlinear viscoelastic response. Usually, LVE regions are small (γc < 10%) for the materials with specific microstructures, such as filled polymers30 and liquid crystalline polymers31 and are very wide (γc > 100%) for some colloidal gels and dilute polymer solutions because of the lack of interactions and delicate microstructures, which are susceptible to large strains.32,33 Therefore, the silica gel matrix obtained by HPGG-catalyzed gelation may be characteristic of specific microstructure and strong interactions, especially in the case of a higher HPGG amount. Insight into the microstructure characteristics of the resulting silica gel was provided when the silica gel was examined with the help of a scanning electron microscope (SEM). Figure 5 shows three SEM pictures for the silica gel matrix containing 10 wt % THEOS and 0.5 wt % HPGG, which were taken with various magnifications. As expected, specific microstructure with two structural elements was found for such silica gel, in which one was a network formed by the crossed or branched macromolecular filaments and another was the silica particles connected to each other. This could be clearly recognized in the SEM pictures taken with higher magnification and had an obvious difference when compared with the morphology of an aerogel sample obtained from aqueous 10 wt % THEOS solution alone by aging for 3 days where only silica particles were observed (Figure 6). It is obvious that added HPGG macromolecules influence the network structure and particle dimension in the silica gel matrix. From Figure 1, we noticed that HPGG is a nonionic polyhydroxy compound. The hydroxyl groups of HPGG could form hydrogen bonds or participate in the condensation reaction with the silanols (Si-OH) produced in the course of hydrolysis of THEOS. As a result, HPGG macromolecules may act as a template for the sol-gel silica matrix generated by THEOS in the HPGG solution. Figure 7 gives a schematic illustration for the formation of the silica gel matrix in the presence of HPGG. Similar microstructure
10668 J. Phys. Chem. B, Vol. 111, No. 36, 2007
Wang and Zhang
Figure 6. The SEM picture of an aerogel sample obtained from aqueous 10 wt % THEOS solution alone.
Figure 7. A schematic illustration for the formation of the silica gel matrix in the presence of HPGG.
Figure 5. The SEM pictures taken with various magnifications for the silica gel matrix containing 10 wt % THEOS and 0.5 wt % HPGG.
characteristics were also observed by us for the hybridized silica gel derived from 10 wt % THEOS and 1.0 wt % chitosan, which was used to construct an amperometric biosensor for hydrogen peroxide.7 When the SEM pictures were compared, however, the network in the case of chitosan became looser and the silica particles became more spherical. This difference may be attributed to the influence of the macromolecule charge on the network structure and particle dimension in the resulting silica matrix because of the fact that chitosan is a polycationic polysaccharide. Among all viscoelastic functions of the gel materials, complex viscosity (η*) is very sensitive to structural changes.34 Figure 8 gives the frequency dependence of η* for all THEOS/HPGG systems investigated, which were measured by dynamic frequency sweep experiments within LVE range. As indicated, these systems exhibited clearly shear-thinning behavior. It has
Figure 8. The frequency dependence of complex viscosity (η*) for the resulting silica gel matrixes. The experiments were carried out at 1.0% strain and 25 °C.
been reported35,36 that some sol-gel silica materials behaved like Newtonian fliuds because of the lack of interactions or entanglements, showing a constant viscosity plateau with changing deformation rate (shear rate or frequency). Apparently, it is not this case here for the THEOS/HPGG gel materials where two kinds of interactions may occur. One belongs to noninteracting components, in which nonionic HPGG macromolecules may be entrapped into the silica matrix and then form interpenetrating networks. Another belongs to interacting components, in which HPGG may form the hydrogen bonds with the silanols resulted from the THEOS hydrolysis and function as a template for the in situ formation of silica gel matrix. In
Sol-Gel Silica Matrix
J. Phys. Chem. B, Vol. 111, No. 36, 2007 10669
Figure 9. Effects of HPGG amount on the storage and loss moduli of the resulting silica gel matrix. The experiments were carried out at 1 Hz, 1.0% strain, and 25 °C.
Figure 11. The relative amounts of vitamin B12 released into a phosphate-buffered saline solution with the pH value of 7.4 from the silica gel matrixes as a function of time (20 °C).
Figure 10. The frequency dependence of storage modulus for the resulting silica gel matrixes. The experiments were carried out at 1.0% strain and 25 °C.
Figure 12. The plot of ln(Mt/Mo) vs ln t for the fitting using Peppas model.
addition, it was also found from Figure 8 that the complex viscosity increased with the increase of HPGG amount at a fixed frequency. For the resultant sol-gel silica matrixes containing HPGG, further investigation was carried out with respect to the effects of HPGG amount on their dynamic moduli at an arbitrary frequency of 1.0 rad/s from a dynamic frequency sweep within the LVE range. As shown in Figure 9, the G′ value was found to increase from 1268 to 3975 Pa while the G′′ value was found to increase from 42 to 147 Pa when the HPGG amount increased from 0.2 to 1.5 wt %. At the same HPGG amount, the G′ value is much greater than the G′′ value, showing a dominant elastic response behavior. In particular, the difference between G′ and G′′ became greater with the increase of HPGG amount. This indicates that an increase of HPGG amount would be favorable for the enhancement of the gel strength. By investigating the storage modulus as a function of angular frequency for the THEOS/HPGG systems containing 10 wt % THEOS and various amounts of HPGG, we found that the storage modulus was almost independent of the frequency for all the systems investigated, as indicated in Figure 10. This phenomenon suggests that the resultant sol-gel silica matrixes in the presence of HPGG possess the behavior of a gel or “solid-like” fluid.37
Drug-Release Behavior. The controlled release of vitamin B12 from the sol-gel silica matrixes resulting from 10 wt % THEOS and various amounts of HPGG was investigated. For a comparison, the pure silica matrix prepared from 10 wt % THEOS alone was also studied with respect to its drug-release behavior. Figure 11 shows the relative amounts of vitamin B12 released into a phosphate-buffered saline solution with the pH value of 7.4 from these silica matrixes as a function of time. It was found that these silica materials have a controlled release action to the model drug. In contrast, the silica matrix containing HPGG suppresses the drug release more obviously. In particular, a strong dependence of the release profile on the amount of introduced HPGG was observed. Moreover, the release rate was found to decrease with the increase of HPGG content. For example, the fraction of drug released after 50 and 180 min was found to be 0.60 and 0.99 for pure silica matrix, 0.42 and 0.76 for the silica matrix containing 0.2 wt % HPGG, 0.40 and 0.73 for the silica matrix containing 0.5 wt % HPGG, 0.36 and 0.64 for the silica matrix containing 1.0 wt % HPGG, and 0.25 and 0.60 for the silica matrix containing 1.5 wt % HPGG, respectively. It may be deduced from these results that the textural properties of the sol-gel silica matrix affect the release of the drug. For the resultant sol-gel silica materials, a dense network structure would slow the release of the loaded drug
10670 J. Phys. Chem. B, Vol. 111, No. 36, 2007
Wang and Zhang
TABLE 1: Kinetic Release Parameters of Vitamin B12 from Various Sol-gel Silica Matrixes and Corresponding Determination Coefficients (R) sample composition
k
n
R
10 wt % THEOS 10 wt % THEOS + 0.2 wt % HPGG 10 wt % THEOS + 0.5 wt % HPGG 10 wt % THEOS + 1.0 wt % HPGG 10 wt % THEOS + 1.5 wt % HPGG
0.044 0.036 0.032 0.035 0.016
0.67 0.64 0.65 0.59 0.69
0.998 0.997 0.996 0.990 0.999
molecules. To investigate the release mechanism, the release curves shown in Figure 9 have been fitted up to a maximum release of 60% with the following equation:38
Mt/Mo ) ktn where Mt/Mo is the fraction of the drug released at time t, k is a constant incorporating structural and geometric feature of the gels, and n is the release exponent which contains the information about the diffusion mechanism. The relationship between ln (Mt/Mo) and ln t is given in Figure 12. The fitting results using the above model are listed in Table 1. The release constant k is found to decrease when THEOS was introduced. In all cases, the release exponent is smaller than 1.0 and greater than 0.5, suggesting a diffusion-controlled release.38 Conclusions A new type of silica gel matrixes were fabricated by solgel process when a completely water-soluble precursor, tetrakis(2-hydroxyethyl)orthosilicates (THEOS), and a completely water-soluble polysaccharide derivative, hydroxypropyl guar gum (HPGG), were used. In the presence of HPGG, the solgel process could proceed rapidly at room temperature without the addition of any organic solvents or the increase of the temperature, which is especially favorable for the encapsulation of drugs or biomolecules. The introduction of HPGG modified the kinetics of sol-gel process and resulted in rapid formation of homogeneous gel matrix. By changing the amount of HPGG introduced, the gelation time for the sol-gel transition and dynamic rheological properties of the resultant gel matrix, as well as the controlled release behavior of the drug from the gel matrix, could be regulated effectively. Acknowledgment. This work was supported by NSFC (20273086; 30470476; 20676155), NSFG (039184; 6023103), Department of Science and Technology of Guangdong Province (2004B33101003), and NCET Program (NCET-04-0810) in Universities of China. References and Notes (1) Ogura, K.; Nakaoka, K.; Nakayama, M.; Kobayashi, M.; Fujii, A. Anal. Chim. Acta 1999, 384, 219-225.
(2) Beakley, L. W.; Yost, S. E.; Cheng, R.; Chandler, B. D. Appl. Catal. A: General 2005, 292, 124-129. (3) Giaume, D.; Buissette, V.; Lahlil, K.; Gacoin, T.; Boilot, J. P.; Casanova, D.; Beaurepaire, E.; Sauviat, M. P.; Alexandrou, A. Prog. Solid State Chem. 2005, 33, 99-106. (4) Miled, O. B.; Ouada, H. B.; Livage, J. Mater. Sci. Eng.: C 2002, 21, 183-188. (5) Zaggout, F. R.; El-Ashgar, N. M.; Zourab, S. M.; El-Nahhal, I. M.; Motaweh, H. Mater. Lett. 2005, 59, 2928-2931. (6) Moleski, R.; Leontidis, E.; Krumeich, F. J. Colloid Interface Sci. 2006, 302, 246-253. (7) Wang, G. H.; Zhang, L. M. J. Phys. Chem. B 2006, 110, 2486424868. (8) Jiang, Y.; Wu, Z.; You, L.; Xiang, H. Colloids Surf. B: Biointerfaces 2006, 49, 55-59. (9) Nicoll, S. B.; Radin, S.; Santos, E. M.; Tuan, R. S.; Ducheyne, P. Biomaterials 1997, 18, 853. (10) Kortesuo, P.; Ahola, M.; Kangas, M.; Leino, T.; Laakso, S.; Vuorilehto, L.; Yli-Urpo, A.; Kiesvaara, J.; Marvola, M. J. Controlled Release 2001, 76, 227. (11) Ahola, M.; Kortesuo, P.; Kangasniemi, I.; Yli-Urpo, A. Int. J. Pharm. 2000, 195, 219. (12) Bottcher, H.; Slowik, P.; Suss, W. J. Sol-Gel Sci. Technol. 1998, 13, 277. (13) Ahola, M.; Sailynoja, E.; Raitavuo, M.; Vaahtio, M.; Salonen, J.; Yli-Urpo, A. Biomaterials 2001, 22, 2163. (14) Fattal, E.; De Rosa, G.; Bochot, A. Int. J. Pharm. 2004, 277, 25. (15) Kortesuo, P.; Ahola, M.; Kangas, M.; Leino, T.; Laakso, S.; Vuorilehto, L.; Yli-Urpo, A.; Kiesvaara, J.; Marvola, M. Int. J. Pharm. 2001, 221, 107. (16) Radin, S.; Ducheyne, P.; Kamplain, T.; Tan, B. H. J. Biomed. Mater. Res. 2001, 57, 313. (17) Aughenbaugh, W.; Radin, S.; Ducheyne, P. J. Biomed. Mater. Res. 2001, 57, 321. (18) Shula, R.; Falaize, S.; Lee, M. H.; Ducheyne, P. Biomaterials 2002, 23, 3113. (19) Bottcher, H.; Jagota, C.; Trepte, J.; Kallies, K. H.; Haufe, H. J. Controlled Release 1999, 60, 57. (20) Wu, Z.; Joo, H.; Lee, T. G.; Lee, K. J. Controlled Release 2005, 104, 497. (21) Kortesuo, P.; Ahola, M.; Karlsson, S.; Kangasniemi, I.; Yli-Urpo, A.; Kiesvaara, J. Biomaterials 2000, 21, 193. (22) Sinha, V. R.; Kumria, R. Int. J. Pharm. 2001, 224, 19. (23) Zhao, S. P.; Ma, D.; Zhang, L. M. Macromol. Biosci. 2006, 6, 445. (24) Mehrotra, R. C.; Narain, R. P. Indian J. Chem. 1966, 5, 444. (25) Billmeye, F. M., Jr. Textbook of Polymer Science; Wiley: New York, 1971. (26) Winter, H. H.; Chambon, F. J. Rheol. 1986, 30, 367. (27) Brinker, C. J.; Scherer, G. W. The Physics and Chemistry of SolGel Processing; Academic Press: Boston, 1990. (28) Shchipunov, Y. A. J. Colloid Interface Sci. 2003, 268, 68. (29) Pierre, A. C. Introduction to Sol-Gel Processing; Kluwer: Boston, 1998. (30) Chow, T. S. J. Phys. Condens. Matter. 1996, 8, 8145. (31) Guskey, S. M.; Winter, H. H. J. Rheol. 1991, 35, 1191. (32) da Silva, M. A.; Areas, E. P. G. J. Colloid Interface Sci. 2005, 289, 394. (33) Hsieh, T. T.; Tiu, C.; Simon, G. P. Polymer 2001, 42, 1931. (34) Wang, L.; Chen, X.; Chen, W. Biomacromolecules 2007, 8, 1109. (35) Zhang, L. M.; Janns, C.; Hsiao, B. S.; Chu, B. J. Colloid Interface Sci. 2003, 266, 339. (36) Ro, J. C.; Chung, I. J. J. Non-Cryst. Solids 1990, 126, 259. (37) Larson R. G. The Structure and Rheology of Complex Fluids; Oxford Univ. Press: New York, 1999. (38) Zhang, L. M.; Wang, G. H.; Lu, H. W.; Yang, C.; Yan, L. J. Bioact. Compat. Polym. 2005, 20, 491.
